Data Visualization - Seaborn Library

Manish Patel

What is Seaborn?

  • Python package for producing statistical graphics.
  • It comes equipped with preset styles and color palettes so you can create complex, aesthetically pleasing charts with a few lines of code.
  • It makes visualization a central part of exploring and understanding data.
  • Its dataset-oriented plotting functions operate on Pandas DataFrames and Numpy arrays containing whole datasets and internally perform the necessary semantic mapping and statistical aggregation to produce informative plots.
  • It is built on top of Matplotlib, but it’s meant to serve as a complement, not a replacement.. Behind the scenes, Seaborn uses matplotlib to draw plots.
  • Seaborn is an important tool used in Exploratory Data Analysis.
%matplotlib inline

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns

Seaborn Settings

  • Matplotlib is highly customizable, but it can be complicated at the same time as it is hard to know what settings to tweak to achieve a good looking plot.
  • Seaborn comes with a number of themes and a high-level interface for controlling the look of matplotlib figures.

The default default seaborn theme, scaling, and color palette is done through the setting:

   sns.set()

It relies on the Matplotlib rcParams customization.

Matplotlib example

def plot_cosine_sine():

    X = np.linspace(-np.pi, np.pi, 256,endpoint=True)
    C = np.cos(X)
    S = np.sin(X)

    # Plot cosine using blue color with a continuous line of width 1 (pixels)
    plt.plot(X, C, color="blue", linewidth=1.0, linestyle="-", label="cos");

    # Plot sine using green color with a continuous line of width 1 (pixels)
    plt.plot(X, S, color="green", linewidth=1.0, linestyle="-", label="sin");

    plt.legend(loc='best')

    # Set x limits
    plt.xlim(-4.0,4.0);

    # Set x ticks
    plt.xticks(np.linspace(-4,4,5,endpoint=True));

    # Set y limits
    plt.ylim(-1.05,1.05);

    # Set y ticks
    plt.yticks(np.linspace(-1,1,5,endpoint=True));

plot_cosine_sine()

Seaborn settings

sns.set()
plot_cosine_sine()

Seaborn Styles

We can also set the background through the function set_style. Some possible settings are:

sns.set_style("dark")
sns.set_style("whitegrid")
sns.set_style("white")

sns.set_style("dark")
plot_cosine_sine()

Seaborn despine

You can use the function despine to remove the spines.

sns.set_style("white")
plot_cosine_sine()
sns.despine()

Using subplot and temporarily setting figure style:

Seaborn subplot

with sns.axes_style("darkgrid"):
   plt.subplot(211)
   plot_cosine_sine()
plt.subplot(212)
plot_cosine_sine()

Identifying Statistical Relationships

  • Statistical analysis is a process of understanding how variables in a dataset relate to each other and how those relationships depend on other variables.
  • When data are visualized properly, the human visual system can see trends and patterns that indicate a relationship.

Load dataset

mpg = sns.load_dataset("mpg")
print(mpg)
      mpg  cylinders  displacement  horsepower  weight  acceleration  \
0    18.0          8         307.0       130.0    3504          12.0   
1    15.0          8         350.0       165.0    3693          11.5   
2    18.0          8         318.0       150.0    3436          11.0   
3    16.0          8         304.0       150.0    3433          12.0   
4    17.0          8         302.0       140.0    3449          10.5   
..    ...        ...           ...         ...     ...           ...   
393  27.0          4         140.0        86.0    2790          15.6   
394  44.0          4          97.0        52.0    2130          24.6   
395  32.0          4         135.0        84.0    2295          11.6   
396  28.0          4         120.0        79.0    2625          18.6   
397  31.0          4         119.0        82.0    2720          19.4   

     model_year  origin                       name  
0            70     usa  chevrolet chevelle malibu  
1            70     usa          buick skylark 320  
2            70     usa         plymouth satellite  
3            70     usa              amc rebel sst  
4            70     usa                ford torino  
..          ...     ...                        ...  
393          82     usa            ford mustang gl  
394          82  europe                  vw pickup  
395          82     usa              dodge rampage  
396          82     usa                ford ranger  
397          82     usa                 chevy s-10  

[398 rows x 9 columns]

Scatterplot

  • The relationship between x and y can be shown for different subsets of the data using the hue, size, and style parameters.
  • These parameters control what visual semantics are used to identify the different subsets.

sns.set()  

sns.scatterplot(x="horsepower", y="mpg", 
                hue="origin", size="weight",  
                sizes=(400, 40), palette="muted",  
                data=mpg);

Lineplot

sns.lineplot(x="horsepower", y="mpg", 
            hue="origin", palette="muted",  
            data=mpg)

Relplot

  • Designed to visualize many different statistical relationships.
sns.set(style="white")  

# Plot miles per gallon against horsepower with other semantics  
sns.relplot(x="horsepower", y="mpg", 
            hue="origin", size="weight",  
            sizes=(400, 40), alpha=.5, palette="muted",  
            height=6, data=mpg)

Manipulating Categorical Data

Categorical scatter

iris = sns.load_dataset("iris")
print(iris)
     sepal_length  sepal_width  petal_length  petal_width    species
0             5.1          3.5           1.4          0.2     setosa
1             4.9          3.0           1.4          0.2     setosa
2             4.7          3.2           1.3          0.2     setosa
3             4.6          3.1           1.5          0.2     setosa
4             5.0          3.6           1.4          0.2     setosa
..            ...          ...           ...          ...        ...
145           6.7          3.0           5.2          2.3  virginica
146           6.3          2.5           5.0          1.9  virginica
147           6.5          3.0           5.2          2.0  virginica
148           6.2          3.4           5.4          2.3  virginica
149           5.9          3.0           5.1          1.8  virginica

[150 rows x 5 columns]

Stripplot

  • Draw a scatterplot where one variable is categorical.
# "Melt" the dataset to "long-form" or "tidy" representation  
iris_m = pd.melt(iris, "species", var_name="measurement")  
  
# Initialize the figure  
f, ax = plt.subplots()  
sns.despine(bottom=True, left=True)  
  
# Show each observation with a scatterplot  
sns.stripplot(x="measurement", y="value", hue="species",  
              data=iris_m, dodge=True, jitter=True,  
              alpha=.25, zorder=1)  
  
# Show the conditional means  
sns.pointplot(x="measurement", y="value", hue="species",  
              data=iris_m, dodge=.532, join=False, palette="dark",  
              markers="d", scale=.75, errorbar=None)  
  
# Improve the legend   
handles, labels = ax.get_legend_handles_labels()  
ax.legend(handles[3:], labels[3:], title="species",  
          handletextpad=0, columnspacing=1,  
          loc="lower right", ncol=3, frameon=True)

Stripplot

Swarmplot

  • Draw a categorical scatterplot with non-overlapping points.
import warnings
warnings.filterwarnings('ignore') # setting ignore as a parameter
sns.set(style="whitegrid", palette="muted") 

# "Melt" the dataset to "long-form" or "tidy" representation  
iris_m = pd.melt(iris, "species", var_name="measurement")  
  
# Draw a categorical scatterplot to show each observation  
sns.swarmplot(x="value", y="measurement", hue="species",  
              palette=["r", "c", "y"], data=iris_m);

Swarmplot

Categorical Distribution Plots

planets = sns.load_dataset("planets")  
print(planets)
               method  number  orbital_period   mass  distance  year
0     Radial Velocity       1      269.300000   7.10     77.40  2006
1     Radial Velocity       1      874.774000   2.21     56.95  2008
2     Radial Velocity       1      763.000000   2.60     19.84  2011
3     Radial Velocity       1      326.030000  19.40    110.62  2007
4     Radial Velocity       1      516.220000  10.50    119.47  2009
...               ...     ...             ...    ...       ...   ...
1030          Transit       1        3.941507    NaN    172.00  2006
1031          Transit       1        2.615864    NaN    148.00  2007
1032          Transit       1        3.191524    NaN    174.00  2007
1033          Transit       1        4.125083    NaN    293.00  2008
1034          Transit       1        4.187757    NaN    260.00  2008

[1035 rows x 6 columns]

BoxPlots

  • Draw a box plot to show distributions with respect to categories.
  • A box plot shows the distribution of quantitative data in a way that facilitates comparisons between variables or across levels of a categorical variable.
  • The box shows the quartiles of the dataset.
sns.set(style="ticks")  
  
# Initialize the figure with a logarithmic x axis  
f, ax = plt.subplots(figsize=(7, 6))  
ax.set_xscale("log")   
  
# Plot the orbital period with horizontal boxes  
sns.boxplot(x="distance", y="method", data=planets,  
            palette="vlag")  
  
# Tweak the visual presentation  
ax.xaxis.grid(True)  
ax.set(ylabel="")  
sns.despine(trim=True, left=True)

BoxPlots

ViolinPlots

  • Draw a combination of boxplot and kernel density estimate.
  • It shows the distribution of quantitative data across several levels of one (or more) categorical variables such that those distributions can be compared.
  • This can be an effective and attractive way to show multiple distributions of data at once.
ax = sns.violinplot(x=planets["distance"])

SubPlots


f, ax = plt.subplots(figsize=(13, 7))

sns.violinplot(x="method", y="distance", data=planets) 
# Finalize the figure  
ax.set(ylim=(-10, 9000))  
sns.despine(left=True, bottom=True)

Brain Network Dataset

df = sns.load_dataset("brain_networks", header=[0, 1, 2], index_col=0)
print(df)
network           1                     2                     3             \
node              1                     1                     1              
hemi             lh         rh         lh         rh         lh         rh   
0         56.055744  92.031036   3.391576  38.659683  26.203819 -49.715569   
1         55.547253  43.690075 -65.495987 -13.974523 -28.274963 -39.050129   
2         60.997768  63.438793 -51.108582 -13.561346 -18.842947  -1.214659   
3         18.514868  12.657158 -34.576603 -32.665958  -7.420454  17.119448   
4         -2.527392 -63.104668 -13.814151 -15.837989 -45.216927   3.483550   
..              ...        ...        ...        ...        ...        ...   
915       -7.429513  -4.813219  -3.670537   1.442261 -19.680431 -29.109356   
916      -33.554138 -38.605621 -25.664803 -30.252352   5.562785  20.182186   
917      -78.539566 -74.197189 -54.041595 -39.970291 -14.277059 -30.606461   
918     -103.235825 -98.744286 -40.109543 -44.907734  12.109148   1.621340   
919      -36.288868 -10.762070 -30.356262 -23.319504  14.252188 -27.559860   

network          4                     5             ...          16  \
node             1                     1             ...           3   
hemi            lh         rh         lh         rh  ...          rh   
0        47.461037  26.746613 -35.898861  -1.889181  ...    0.607904   
1        -1.210660 -19.012897  19.568010  15.902983  ...   57.495071   
2       -65.575806 -85.777428  19.247454  37.209419  ...   28.317369   
3       -41.800869 -58.610184  32.896915  11.199619  ...   71.439629   
4       -62.613335 -49.076508  18.396759   3.219077  ...   95.597565   
..             ...        ...        ...        ...  ...         ...   
915      -3.376562  -8.764893   0.017912 -25.305094  ...   58.285793   
916      17.911247  24.653582 -32.935612 -21.783203  ...   59.864819   
917      60.628521  62.027023 -32.800556 -37.021500  ...   68.767868   
918      33.765560  55.356071 -14.330512 -17.224781  ...  103.155251   
919      17.279512 -19.060152  26.558777  21.377319  ...  102.086304   

network                                 17                                   \
node              4                      1                     2              
hemi             lh          rh         lh         rh         lh         rh   
0        -70.270546   77.365776 -21.734550   1.028253   7.791784  68.903725   
1        -76.393219  127.261360 -13.035799  46.381824 -15.752450  31.000332   
2          9.063977   45.493263  26.033442  34.212200   1.326110 -22.580757   
3         65.842979  -10.697547  55.297466   4.255006  -2.420144  12.098393   
4         50.960453  -23.197300  43.067562  52.219875  28.232882 -11.719750   
..              ...         ...        ...        ...        ...        ...   
915       68.557411   22.833048  76.179489  51.934669  -6.614513  -6.690762   
916       23.004578    7.657463  50.962399  13.696922  63.503616  57.401176   
917       49.664017   64.551498  43.800747   8.144480  47.281460  70.499649   
918      106.454849    9.046827  46.674419  40.954796   0.877180  37.577152   
919       80.051140   -2.642610 -12.229620  -6.596726  17.665163  16.153173   

network                                    
node             3                      4  
hemi            lh          rh         lh  
0       -10.520872  120.490463 -39.686432  
1       -39.607521   24.764011 -36.771008  
2        12.985169  -75.027451   6.434262  
3       -15.819172  -37.361431  -4.650954  
4         5.453649    5.169828  87.809135  
..             ...         ...        ...  
915      22.893030   48.274380  76.228455  
916      24.974548   51.972153  64.538788  
917      66.994400   81.539246  64.969772  
918      20.517746    3.124434  56.718388  
919       8.300399   33.687531  17.960655  

[920 rows x 62 columns]

Correlation Plots

sns.set(style="whitegrid")  
  
# Load the example dataset of brain network correlations  
df = sns.load_dataset("brain_networks", header=[0, 1, 2], index_col=0)  
  
# Pull out a specific subset of networks  
used_networks = [1, 3, 4, 5, 6, 7, 8, 11, 12, 13, 16, 17]  
used_columns = (df.columns.get_level_values("network")  
                          .astype(float)  
                          .isin(used_networks))  
df = df.loc[:, used_columns]  
  
# Compute the correlation matrix and average over networks  
corr_df = df.corr().groupby(level="network").mean()  
corr_df.index = corr_df.index.astype(int)  
corr_df = corr_df.sort_index().T  
  
# Set up the matplotlib figure  
f, ax = plt.subplots(figsize=(11, 6))  
  
# Draw a violinplot with a narrower bandwidth than the default  
sns.violinplot(data=corr_df, palette="Set3", bw=1, cut=.2, linewidth=1)  
  
# Finalize the figure  
ax.set(ylim=(-.7, 1.05))  
sns.despine(left=True, bottom=True)

Correlation Plots

BoxenPlots

  • Draw an enhanced box plot for larger datasets.
  • This style of plot was originally named a “letter value” plot because it shows a large number of quantiles that are defined as “letter values”.
  • It is similar to a box plot in plotting a nonparametric representation of a distribution in which all features correspond to actual observations.
  • By plotting more quantiles, it provides more information about the shape of the distribution, particularly in the tails.
ax = sns.boxenplot(x=planets["distance"])

SubPlots

f, ax = plt.subplots(figsize=(13, 7))

sns.boxenplot(x="method", y="distance", data=planets) 
# Finalize the figure  
ax.set(ylim=(-10, 9000))  
sns.despine(left=True, bottom=True)

Categorical Estimate Plots

titanic = sns.load_dataset("titanic")
print(titanic)
     survived  pclass     sex   age  sibsp  parch     fare embarked   class  \
0           0       3    male  22.0      1      0   7.2500        S   Third   
1           1       1  female  38.0      1      0  71.2833        C   First   
2           1       3  female  26.0      0      0   7.9250        S   Third   
3           1       1  female  35.0      1      0  53.1000        S   First   
4           0       3    male  35.0      0      0   8.0500        S   Third   
..        ...     ...     ...   ...    ...    ...      ...      ...     ...   
886         0       2    male  27.0      0      0  13.0000        S  Second   
887         1       1  female  19.0      0      0  30.0000        S   First   
888         0       3  female   NaN      1      2  23.4500        S   Third   
889         1       1    male  26.0      0      0  30.0000        C   First   
890         0       3    male  32.0      0      0   7.7500        Q   Third   

       who  adult_male deck  embark_town alive  alone  
0      man        True  NaN  Southampton    no  False  
1    woman       False    C    Cherbourg   yes  False  
2    woman       False  NaN  Southampton   yes   True  
3    woman       False    C  Southampton   yes  False  
4      man        True  NaN  Southampton    no   True  
..     ...         ...  ...          ...   ...    ...  
886    man        True  NaN  Southampton    no   True  
887  woman       False    B  Southampton   yes   True  
888  woman       False  NaN  Southampton    no  False  
889    man        True    C    Cherbourg   yes   True  
890    man        True  NaN   Queenstown    no   True  

[891 rows x 15 columns]

PointPlots

  • Show point estimates and confidence intervals using scatter plot glyphs.
  • A point plot represents an estimate of central tendency for a numeric variable by the position of scatter plot points and provides some indication of the uncertainty around that estimate using error bars.
  • Point plots can be more useful than bar plots for focusing comparisons between different levels of one or more categorical variables.
ax = sns.pointplot(x="class", y="survived", hue="who", 
                   data=titanic)

CountPlots

  • Show the counts of observations in each categorical bin using bars.
  • A count plot can be thought of as a histogram across a categorical, instead of quantitative, variable.
ax = sns.countplot(x="class", hue="who", data=titanic)

CatPlot

g = sns.catplot(x="class", hue="who", col="survived",
                data=titanic, kind="count",
                height=4, aspect=.7);

BarPlot

  • Show point estimates and confidence intervals as rectangular bars.
  • A bar plot represents an estimate of central tendency for a numeric variable with the height of each rectangle and provides some indication of the uncertainty around that estimate using error bars.
ax = sns.barplot(x="class", y="survived", data=titanic)

Hue

ax = sns.barplot(x="class", y="survived", hue="who", data=titanic)

Visualizing the Distribution of a Dataset

DistPlots

  • Flexibly plot a univariate distribution of observations.
sns.set(style="white", palette="muted", color_codes=True)  
rs = np.random.RandomState(10)  
  
# Set up the matplotlib figure  
f, axes = plt.subplots(2, 2, figsize=(7, 7), sharex=True)  
sns.despine(left=True)  
  
# Generate a random univariate dataset  
d = rs.normal(size=100)  
  
# Plot a simple histogram with binsize determined automatically  
sns.distplot(d, kde=False, color="b", ax=axes[0, 0])  
  
# Plot a kernel density estimate and rug plot  
sns.distplot(d, hist=False, rug=True, color="r", ax=axes[0, 1])  
  
# Plot a filled kernel density estimate  
sns.distplot(d, hist=False, color="g", kde_kws={"shade": True}, ax=axes[1, 0])  
  
# Plot a historgram and kernel density estimate  
sns.distplot(d, color="m", ax=axes[1, 1])  
  
plt.setp(axes, yticks=[])  
plt.tight_layout()

DistPlots

KDEPlot

  • Fit and plot a univariate or bivariate kernel density estimate.
tips = sns.load_dataset("tips")

sns.kdeplot(
   data=tips, x="total_bill", hue="size",
   fill=True, common_norm=False, palette="crest",
   alpha=.5, linewidth=0,
)

KDEPlot

PairPlots

  • Plot pairwise relationships in a dataset.
sns.set(style="ticks")  
#iris = sns.load_dataset("iris") 
sns.pairplot(iris, hue="species");

PairPlots

Visualizing Linear Relationships

RegPlots

  • Plot data and a linear regression model fit.
g = sns.regplot(x="sepal_length", y="sepal_width",  
                 data=iris)

LMPlot

  • Enhance a scatterplot to include a linear regression model (and its uncertainty).
iris = sns.load_dataset("iris")

# Plot sepal with as a function of sepal_length across days  
g = sns.lmplot(x="sepal_length", y="sepal_width", hue="species",  
               truncate=True, height=5, data=iris)  
  
# Use more informative axis labels than are provided by default  
g.set_axis_labels("Sepal length (mm)", "Sepal width (mm)");

JointPlots

  • Draw a plot of two variables with bivariate and univariate graphs.
g = sns.jointplot(x="sepal_length", y="sepal_width", data=iris)

Variation

g = sns.jointplot(x="sepal_length", y="sepal_width", data=iris, 
                  kind="reg", space=0, color="g")

Useful Functions

Heatmap

  • Plot rectangular data as a color-encoded matrix.
uniform_data = np.random.rand(10, 12)
ax = sns.heatmap(uniform_data)

Flight dataset heatmap

sns.set()  
  
# Load the example flights dataset and convert to long-form  
flights_long = sns.load_dataset("flights")  
flights = flights_long.pivot("month", "year", "passengers")  
  
# Draw a heatmap with the numeric values in each cell  
f, ax = plt.subplots(figsize=(11, 8))  
sns.heatmap(flights, annot=True, fmt="d", linewidths=.5, ax=ax);

Flight dataset heatmap

Clustermap

  • Plot a matrix dataset as a hierarchically-clustered heatmap.
species = iris.pop("species")
g = sns.clustermap(iris)

Manish Patel

Clustermap